Dual Authentication Voting System with Facial Recognition and Fingerprint Using CNN
P. Asha, V Vidhya, B. Ankayarkanni, Vasantha Sunil Reddy, J. Albert Mayan, D. Usha Nandini
Abstract
Voting is one of the notably common and basic right of all citizens of the country. In the latest loksabha election 2019, conducted in India, only 67.1% vote was being recorded. Conducting election in a democratic country like India, have major challenges of security threads, false voting, manipulation in recording vote and authentication of voter. Our country India employs the EVM (Electronic Voting Machine) since 1982. The system also has its flaws as vote rigging is a possibility. Hence, in this paper we give a solution to the voting problem. The data for the project can be acquired from the dataset of Aadhar card that contain the photo. By using CNN (Convolutional Neural Network) the face of the voter is detected and compared to the Aadhar dataset. This will be the first step of the authentication. Then the vote is casted through fingerprint, which is sensed through sensor and is authenticated. This will allow the voter to vote only once and to only one person. The recorded vote is saved in the cloud server and is counted immediately. This makes the system more efficient and authenticated and avoids rigging of vote, than the existing system of voting.